Defensive Refusal Bias: How Safety Alignment Fails Cyber Defenders

Documenti analoghi

• PropensityBench: Evaluating Latent Safety Risks in Large Language Models via an Agentic Approach
  di: Sehwag, Udari Madhushani, et al.
  Pubblicazione: (2025)
• AdvBDGen: Adversarially Fortified Prompt-Specific Fuzzy Backdoor Generator Against LLM Alignment
  di: Pathmanathan, Pankayaraj, et al.
  Pubblicazione: (2024)
• In-Context Learning with Topological Information for Knowledge Graph Completion
  di: Sehwag, Udari Madhushani, et al.
  Pubblicazione: (2024)
• ROK-FORTRESS: Measuring the Effect of Geopolitical Transcreation for National Security and Public Safety
  di: Lee, Michael S., et al.
  Pubblicazione: (2026)
• GenARM: Reward Guided Generation with Autoregressive Reward Model for Test-time Alignment
  di: Xu, Yuancheng, et al.
  Pubblicazione: (2024)